checkpoint-412/trainer_state.json

{
  "best_global_step": 400,
  "best_metric": 0.29742929339408875,
  "best_model_checkpoint": "codellama_utests_embedded_v2/checkpoint-400",
  "epoch": 4.0,
  "eval_steps": 200,
  "global_step": 412,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4854368932038835,
      "grad_norm": 0.3694750666618347,
      "learning_rate": 9.8e-05,
      "loss": 2.0288,
      "step": 50
    },
    {
      "epoch": 0.970873786407767,
      "grad_norm": 0.13560760021209717,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.4491,
      "step": 100
    },
    {
      "epoch": 1.4563106796116505,
      "grad_norm": 0.08724918961524963,
      "learning_rate": 0.0001685897435897436,
      "loss": 0.3395,
      "step": 150
    },
    {
      "epoch": 1.941747572815534,
      "grad_norm": 0.07571534067392349,
      "learning_rate": 0.00013653846153846154,
      "loss": 0.3168,
      "step": 200
    },
    {
      "epoch": 1.941747572815534,
      "eval_loss": 0.3129767179489136,
      "eval_runtime": 154.0926,
      "eval_samples_per_second": 1.343,
      "eval_steps_per_second": 0.169,
      "step": 200
    },
    {
      "epoch": 2.4271844660194173,
      "grad_norm": 0.07541365176439285,
      "learning_rate": 0.0001044871794871795,
      "loss": 0.3,
      "step": 250
    },
    {
      "epoch": 2.912621359223301,
      "grad_norm": 0.08043674379587173,
      "learning_rate": 7.243589743589744e-05,
      "loss": 0.3006,
      "step": 300
    },
    {
      "epoch": 3.3980582524271843,
      "grad_norm": 0.08717360347509384,
      "learning_rate": 4.038461538461539e-05,
      "loss": 0.2905,
      "step": 350
    },
    {
      "epoch": 3.883495145631068,
      "grad_norm": 0.07547729462385178,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.2857,
      "step": 400
    },
    {
      "epoch": 3.883495145631068,
      "eval_loss": 0.29742929339408875,
      "eval_runtime": 154.1463,
      "eval_samples_per_second": 1.343,
      "eval_steps_per_second": 0.169,
      "step": 400
    }
  ],
  "logging_steps": 50,
  "max_steps": 412,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.406536142516388e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}